Skip to content

sint-ai/sint-protocol

AI Agents & Assistants

A governance layer that enforces security, permissions, and economic controls for AI agents interacting with physical systems like robots, drones, and smart‑home devices.

TypeScript Latest v0.2.6 · 1mo ago Security brief →

Features

  • Enforces capability‑based permissions on every tool call, robot command, or actuator movement
  • Provides tiered human oversight (T0–T3) for graduated approval of actions
  • Generates tamper‑evident, SHA‑256 hash‑chained audit logs and attestation receipts
  • Supports multiple physical‑world bridges (ROS 2, MAVLink, industrial protocols)
  • Offers economic routing, budgets, and budget‑aware action execution

Recent releases

View all 6 releases →
No immediate action
v0.2.6 New feature

Docker independence, schema tightenings, inputSchema

No immediate action
v0.2.4 New feature

Clean install command

No immediate action
v0.2.3 New feature

Install instructions

No immediate action
v0.2.0 New feature

sint-scan + Policy Bundle Spec

No immediate action
v0.1.0 New feature

sint-scan + RFC-001

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
9
Forks
7
Languages
TypeScript Python JavaScript
Downloads/week
3 ↓60%
NPM Maintainers
1
Contributors
13
TypeScript
Types included ✓

Install & Platforms

Install via
npm docker

Alternative to

Microsoft AGT MCP Baseline SROS2

Beta — feedback welcome: [email protected]